From the course: Defending and Deploying AI by Pearson
Unlock this course with a free trial
Join today to access over 25,200 courses taught by industry experts.
Introducing retrieval-augmented generation (RAG)
From the course: Defending and Deploying AI by Pearson
Introducing retrieval-augmented generation (RAG)
One of the most popular topics nowadays in AI is Retrieval Augmented Generation, and I would like to actually go over what it is, what it entails, and what are the advantages that you have with Retrieval Augmented Generation. What you're seeing in front of you, of course, is one of my articles in my personal blog that we just mentioned earlier. This one specifically is around line chain. And I wrote it several months ago. However, what I would like to do is take advantage of this diagram here. So let me actually start by defining what is RAG, or Retrieval Augmented Generation. So RAG, or Retrieval Augmented Generation, is basically a machine learning and AI concept that aims to enhance the capabilities of Gen. AI models with external knowledge sourced from either a document collection, another database, and so on, and basically act as a framework that is aimed to enhance the quality of the responses, basically for you to get better answers and better output of the models and reduce…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.
Contents
-
-
Learning objectives2m 54s
-
Overview of the evolution of AI-driven tools4m 32s
-
Large language models (LLMs) and small language models (SLMs)8m 32s
-
Introducing LangChain, LangGraph, Llama Index, and other orchestration frameworks6m 7s
-
An overview of open-source AI models and Hugging Face3m 11s
-
Introducing retrieval augmented generation (RAG)11m 29s
-
Introducing embedding models5m 45s
-
(Locked)
Introducing vector databases: pgvector, Chroma, MongoDB Atlas Vector Search, and others3m 42s
-
(Locked)
Introducing semantic search4m 30s
-
-
-
(Locked)
Learning objectives2m 11s
-
(Locked)
Exploring how AI is revolutionizing software development2m 27s
-
(Locked)
Surveying GitHub Copilot, Cursor, and Cody2m 27s
-
(Locked)
Exploring the CODEX model1m 18s
-
(Locked)
Generating code from a prompt5m 34s
-
(Locked)
Explaining existing code2m 55s
-
(Locked)
Generating comments3m 52s
-
(Locked)
Prompt engineering for software development8m 46s
-
(Locked)
-
-
(Locked)
Learning objectives1m 18s
-
(Locked)
Understanding the significance of LLMs in the AI landscape7m 6s
-
(Locked)
Exploring the resources for this course: GitHub repositories and others2m 54s
-
(Locked)
Introducing retrieval-augmented generation (RAG)12m 24s
-
(Locked)
Understanding the OWASP Top 10 risks for LLMs5m 46s
-
(Locked)
Exploring the MITRE ATLAS™ (Adversarial Threat Landscape for Artificial-Intelligence Systems) framework5m 38s
-
(Locked)
Understanding the NIST taxonomy and terminology of attacks and mitigations7m 8s
-
(Locked)
-
-
(Locked)
Learning objectives1m 1s
-
(Locked)
Defining prompt injection attacks11m 41s
-
(Locked)
Exploring real-life prompt injection attacks3m 57s
-
(Locked)
Using ChatML for OpenAI API calls to indicate to the LLM the source of prompt input10m 4s
-
(Locked)
Enforcing privilege control on LLM access to back-end systems6m 10s
-
(Locked)
Best practices around API tokens for plugins, data access, and function-level permissions3m 2s
-
(Locked)
Understanding insecure output handling attacks3m 22s
-
(Locked)
Using the OWASP ASVS to protect against insecure output handling4m 43s
-
(Locked)
-
-
(Locked)
Learning objectives46s
-
(Locked)
Understanding training data poisoning attacks4m 26s
-
(Locked)
Exploring model denial of service attacks3m 11s
-
(Locked)
Understanding the risks of the AI and ML supply chain8m 33s
-
(Locked)
Best practices when using open-source models from Hugging Face and other sources12m 45s
-
(Locked)
Securing Amazon Bedrock, SageMaker, Microsoft Azure AI Services, and other environments16m 5s
-
(Locked)
-
-
(Locked)
Learning objectives1m 27s
-
(Locked)
Overview of AI labs and sandboxes: Home-based vs. cloud-based7m 4s
-
(Locked)
Choosing the right hardware: GPUs, CPUs, and memory10m 9s
-
(Locked)
Building or buying prebuilt systems3m 53s
-
(Locked)
Choosing the operating system: Linux, Windows, and macOS4m 21s
-
(Locked)
Surveying essential software: Python, Anaconda, Jupyter Notebooks, and other frameworks4m 41s
-
(Locked)
Introducing Hugging Face3m 42s
-
(Locked)
Introducing Ollama3m 38s
-
(Locked)
Installing Ollama6m 24s
-
(Locked)
Ollama integrations14m 42s
-
(Locked)
Exploring the Ollama REST API3m 3s
-
(Locked)
Exploring RAG5m 24s
-
(Locked)
Leveraging RAGFlow13m 56s
-
(Locked)
-
-
(Locked)
Learning objectives38s
-
(Locked)
Advantages and disadvantages of cloud AI labs and sandboxes8m 18s
-
(Locked)
Introducing Amazon Bedrock6m 59s
-
(Locked)
Surveying Amazon SageMaker12m 56s
-
(Locked)
Exploring Google Vertex AI14m 13s
-
(Locked)
Using Microsoft Azure AI Foundry10m 11s
-
(Locked)
Discussing cost management and security6m 45s
-
(Locked)
Using Terraform to deploy Ollama in the cloud3m 28s
-
(Locked)
-
-
(Locked)
Learning objectives43s
-
(Locked)
Using hybrid AI labs: combining home and cloud resources1m 57s
-
(Locked)
Synchronizing data and projects4m 16s
-
(Locked)
Leveraging the strengths of both environments5m 40s
-
(Locked)
Running open-source models available on Hugging Face4m 24s
-
(Locked)
Introducing LangChain3m 14s
-
(Locked)
Introducing LlamaIndex2m 12s
-
(Locked)
Understanding embedding models6m 23s
-
(Locked)
Using vector databases4m 56s
-
(Locked)
-
-
(Locked)
Learning objectives1m 30s
-
(Locked)
The essentials of RAG7m 21s
-
(Locked)
Exploring the GitHub repositories and additional resources3m 40s
-
(Locked)
Embeddings and embedding models18m 5s
-
(Locked)
Indexing techniques13m 20s
-
(Locked)
Vector databases6m 32s
-
(Locked)
Chunking strategies9m 47s
-
(Locked)
RAG vs. fine-tuning14m 2s
-
(Locked)
RAG, RAG Fusion, and RAPTOR11m 51s
-
(Locked)
Running open-weight models with Ollama17m
-
(Locked)
Exploring Open WebUI and other Ollama plugins12m 47s
-
(Locked)
Introduction to AI agents and agentic implementations6m 54s
-
(Locked)
Introduction to agentic RAG6m 6s
-
(Locked)
Introducing the Model Context Protocol (MCP)14m 19s
-
(Locked)
Introducing A2A and AGNTCY20m 57s
-
(Locked)
-
-
(Locked)
Learning objectives1m 7s
-
(Locked)
Mastering prompt engineering4m 14s
-
(Locked)
Exploring basic prompt chain examples10m 3s
-
(Locked)
Creating prompt branching chains12m 32s
-
(Locked)
Exploring parallel prompt chains7m 37s
-
(Locked)
Creating a basic RAG application15m 25s
-
(Locked)
Creating a complete RAG application4m 15s
-
(Locked)
-
-
(Locked)
Learning objectives1m 3s
-
(Locked)
Introduction to AI agent frameworks6m 18s
-
(Locked)
Surveying CrewAI8m 39s
-
(Locked)
Introducing LangGraph2m 41s
-
(Locked)
Exploring examples of LangGraph in action16m 37s
-
(Locked)
Exploring an example of agents with MCP servers13m 44s
-
(Locked)
Securing agentic implementations12m 4s
-
(Locked)